How To Deal With Persistent Storage (e.g. Databases) In Docker?

The best way to deal with persistent data storage (such as a database) in Docker is to use Docker’s volume API (for docker 1.9.0 or newer) or use data-only containers for older versions of Docker.

Docker 1.9.0 or newer

The best approach for you is to use docker volume API as shown in the following example:

Copied!

docker volume create --name myvolume
docker run -d -v myvolume:/container/path/for/volume container_image my_command

Let’s break down the example command:

The first command creates a new volume named myvolume in the current working directory
The second command runs the container image container_image and configures it to consume the newly created volume myvolume

If you create a container with a -v volume_name:/container/fs/path Docker will automatically create a named volume for you that can:

Be listed through the docker volume ls
Be identified through the docker volume inspect volume_name
Backed up as a regular directory
Backed up as before through a --volumes-from connection

You can also identify dangling volumes (unused volumes)

Copied!

docker volume ls -f dangling=true

And then remove them using their name:

Copied!

docker volume rm <volume_name>

Docker 1.8.X and older

The approach that seems to work best for production is to use a data-only container. The data-only container is run on a barebones image and does nothing except expose a data volume.

Then you can run any other container to have access to the data container volumes:

Copied!

docker run --volumes-from data_container some_other_container command_to_exec

You can read more about this technique in the official documentation.

Got an article suggestion? Let us know

Explore more

Docker

Solved: Cron missing newline before EOF

This error may also happen when a crontab file is generated automatically and the generator failed to insert a newline character at the end of the crontab file.